Editor in Chief
Imagine clusters and supercomputers coming together and creating a child with Cray as the delivery doctor. Last night at the SC18, Cray Inc. unveiled its next generation exascale supercomputer – Shasta.
This supercomputer will provide scale-able solutions for today’s challenges and meet future computational needs.
In a world with increasing challenges and dwindling resources, the need to solve the next generations problems at an increasing speed is pertinent.
With exascale computing, problems such as cancer research and climate control can be solved faster than before. The need for exascale computing arose from the national strategic computing initiative (NSCI).
This initiative provides a framework for how to handle the exascale computing project-the specific means by which we get an exascale computer, the exascale computing initiative -using large scale simulation resources in places where they have not been used before and moving beyond Moore’s law, aiming to answer how we envision those elements of computing that go beyond the current generation.
With the Cray system, comes a brand-new interconnect – slingshot.
Shasta is a family of systems that will be rolling out over the next several years. It is an entirely new design and is set to be the technology to underpin the next era of supercomputing, characterized by exascale performance capability, new data-centric workloads, and heterogenous computing. It will be characterized by a software rack, brand new interconnect called slingshot and a variety of a couple different ways of packaging. A packaging so unique you would say, that’s a Cray! Such as a 19-inch rack to take our interconnect into standard packaging.
Shasta was designed to meet the demanding and evolving needs of the exascale era. Working hand-in-hand with academia, Shasta represents an interesting opportunity for academia by providing platforms to start small and scale, utilizing its new interconnect for the most data centric applications and for use in advanced research with varying types of workload whether it is simulation, AI, big data analytics or even graph analytics.
“It’s a cost efficient, scale efficient type of solution” said Frederick Kohout, Senior Vice President and Chief Marketing Officer, one that can grow with the needs of the organization. Shasta will need the right processor-partner to break the Top500 record and Cray hopes that will do the trick.